Download Stylised Facts for New Zealand Business Cycles

7 June 2012
Work in progress
Comments welcome
Stylised Facts for New Zealand Business Cycles:
A Post-1987 Perspective
Stuart McKelviea and Viv B Halla*
a
School of Economics and Finance, Victoria University of Wellington
Abstract
Key features of NZ business cycles were established for the period 1966q4 to 1990q1 by Kim,
Buckle and Hall (1994) (KBH), but the conduct of fiscal, monetary and labour market policy and
the behaviour of New Zealand‟s economy have changed considerably since then. Our results for
the period 1987q2 to 2010q4 show a reduction in volatility and a rise in persistence for both the
real economy and for price and monetary variables. Government sector, open economy,
monetary and labour market results differ from those advanced in KBH. Overall, we establish a
more credible set of benchmark regularities, to help underpin the construction and use of
contemporary NZ macroeconomic models.
JEL Classification: E32, C54
Keywords: Business cycle stylised facts; New Zealand; Growth cycle analysis
___________________________
* Corresponding author. Email: [email protected]
1
Stylised Facts for New Zealand Business Cycles:
A Post-1987 Perspective
1
Introduction
Over the past 60 years or so, the New Zealand economy and the international environment in
which it operates have undergone major changes. Key features of New Zealand business cycles were
established for the period 1966q4 to 1990q1 by Kim, Buckle, and Hall (1994) (KBH)1, but a number of
financial, fiscal, external sector, and labour market regularities had to remain unresolved in that study.
Moreover, the conduct of fiscal, monetary and labour market policy in New Zealand and the behaviour of
its economy have changed considerably since then. It is timely therefore to present a more contemporary
set of stylised facts for New Zealand business cycles.
We use well-established univariate and bivariate growth cycle methodology to quantify volatility,
persistence, and co-movements of aggregate output with the economy‟s salient macroeconomic variables,
for the period from 1987q2 to 2010q4. Unlike many other comparable studies, we give particular
attention to the extent to which volatility, and contemporaneous and non-contemporaneous crosscorrelations have moved within our sample period2. Results are assessed in the context of KBH (1994),
McCaw (2007), and Hall and McDermott (2011) (HMcD).
The stylised facts approach can be traced back at least as far as Lucas (1997, p 10) who stated
that “with respect to the qualitative behaviour of co-movements among [economic time] series, business
cycles are all alike”. This claim implied the possibility of a unified explanation of business cycles grounded
in the general laws governing market economies. The insights of Lucas led to increased emphasis on
business cycle facts being the starting point for any macroeconomic theory and model which purported to
explain and predict aggregate economic activity. So much so that nowadays, „stylised facts‟ provide not
only one important litmus test for existing macroeconomic models, but also a guide to the construction
of new models.
In the late 1980s and early 1990s, a series of papers revisited the stylised facts for a number of
industrialised economies. These studies included, influentially, Greenwald, Stiglitz, Hall, and Fischer
(1988), Kydland and Prescott (1990), Backus and Kehoe (1992), and Backus, Kehoe, and Kydland (1994),
among others. For some macroeconomic variables, these studies established consensus, for others
diversity. Unfortunately, very few of these findings were directly portable to small open economies such
as New Zealand and Australia, contrary to the general view expressed by Lucas (1977) that business cycles
are all alike. For New Zealand, this limitation was responded to initially by Kim, Buckle and Hall (1994),
1
Prior to this study, there had been no systematic analysis of the properties of New Zealand business cycles since
the work by Haywood (1972) and Haywood and Campbell (1976).
2
Our results come from programs written in RATS by Dr Kunhong Kim.
2
and for Australia by Crosby and Otto (1995) and by Fisher, Otto and Voss (1996). More recent results
for Australia have been reported in Tawardros (2011).
In their examinations, KBH (1994) documented the volatility, autocorrelation, and crosscorrelation with real GDP of a comprehensive set of macroeconomic variables. They established a
number of key features of New Zealand growth cycles, including the exceptional volatility of the
economy; the procyclicality of real variables, the countercyclical tendency of price fluctuations, and a
number of robust labour market regularities. However, the scale of changes to the economy in the latter
part of their sample period frustrated efforts to obtain credible regularities with respect to New Zealand‟s
financial, fiscal, and open economy variables. Clarification of the uncertainties in these areas provided a
major motivation for this study.
Since the KBH study, research into stylised facts of New Zealand business cycles has been
limited. Sanyal and Ward (1995) investigated a number of bivariate regularities, although their analysis is
preoccupied with the structural disturbances of the 1970s. McCaw (2007) utilised New Zealand‟s updated
chain-volume data from 1987(2) to 2006(2) to examine relationships for an exhaustive set of
macroeconomic variables. But the reporting of cross-correlation results for not only New Zealand‟s
output gap, but also for its CPI and nontradables inflation gaps, and the 90-day interest rate gap, together
with cross correlations of key New Zealand variables with those for Australia and the U.S. meant that the
study did not have a primary focus on the fiscal, financial, and external sector uncertainties identified in
KBH. Nonetheless, where relevant, we compare our findings with those of McCaw. Similarly where
relevant, our results are compared with those reported for New Zealand unobserved components
business cycles in Hall and McDermott (2011, Table 7) for the sample period 1983q2 to 2006q4.
The statistical results we report are motivated by the following key questions: To what extent
have the volatility and persistence of New Zealand‟s business cycles decreased since KBH? Has real
variable regularity been maintained? Does government expenditure exhibit a pro-, counter-, or acyclical
relationship with output? Can one establish credible regularities for other fiscal variables, including for
taxation, transfers and a public debt to GDP ratio? Has more recent data helped clarify the cyclical
relationship between net exports and real output? Can one establish a sufficiently clear role for the termsof-trade and the real aggregate exchange rate? Have a new set of monetary regularities emerged in the
post-reform period? And how, if at all, have labour market regularities changed?
Our univariate and bivariate methodology is described succinctly in Section 2. Our stylised facts
for New Zealand business cycles are reported and commented on in Section 3. Section 4 concludes.
2
Methodology
As documented in Appendix C, our raw data series have been sourced from Statistics New
Zealand (SNZ), the Reserve Bank of New Zealand (RBNZ), and the New Zealand Treasury. Variables
3
from the latter data set have been utilised previously in Claus et al (2006), Dungey and Fry (2009) and
Hall and McDermott (2011). Our sample size is restricted to the period from 1987q2 to 2010q4, to
coincide with the start point of SNZ‟s chain-volume series based on the International System of National
Accounts 1993.
The set of variables selected for analysis was largely guided by stylised facts which were either
unresolved in KBH or may have changed significantly since then. This necessitated considering real GDP
expenditure components, external variables, monetary variables, and labour market variables. Series were
seasonally adjusted as required and then log transformed, with the exception of those containing negative
observations (e.g. net exports) or those already expressed as a percentage (e.g. interest rates).
To achieve consistency with previous studies for New Zealand, the methodology of this paper is
basically as utilised in KBH (1994) and McCaw (2007). This meant the adoption of growth cycle rather
than classical cycle methodology, and as is well-known immediately raises the controversial issue of
detrending.
An assessment of fluctuations about a trend in an observed economic variable ( ), requires
decomposition of its time series into a non-stationary trend component ( ) and a stationary cyclical
component
:
A considerable number of de-trending procedures are possible, but we utilise the well-known
algorithm developed by Hodrick & Prescott (1980) (HP). This HP filter has the considerable advantage of
providing a simple, intuitive, and highly operational method of trend extraction. It generates an estimate
of the secular and cyclical components according to the convex minimization problem
where
. Implicit in the above equation is a trade-off between the degree of fit and the smoothness
of the trend component. The first term in the equation represents the sum of the squared deviations from
trend and penalises the cyclical component. The second term is the sum of the squares of the trend
component‟s second differences, and penalises variations in the trend growth rate. The arbitrarily chosen
smoothing parameter, , determines the rate at which the latter is penalised relative to the former. As
approaches zero, the trend component conforms more closely to the actual data series; while as
to
tends
, the trend component becomes linear. We adopt Hodrick and Prescott‟s (1980) recommended
smoothing parameter of 1600 for quarterly data, as previously used successfully for New Zealand series
reported in KBH (1994) and in Hall, Kim, and Buckle (1998). Adoption of this parameter value also
4
facilitates direct comparisons with results from KBH (1994), McCaw (2007), and studies for many other
countries.
However, use of the HP filter is not without controversy. In particular, it is argued that use of the
standardised smoothing parameter has a tendency to generate spurious correlations by attributing
medium to long-run variations in the data to the cycle when in fact, they are part of the trend (Harvey and
Jaeger, 1993). HP filters are also criticised for producing unusual cyclical components toward either end
of the sample (Baxter & King, 1999). Such criticism provokes the consideration of alternative detrending
methodologies. Prominent alternatives include the band-pass filter of Baxter and King (1999) (BK) and
the unobserved components (UC) model of Harvey (1985, 1989) . For example, Choy (2011) has recently
used BK and UC methodology to assess stylised facts for Singaporean business cycles, and Tawardros
(2011) has used HP and UC methodology to re-examine Australia‟s stylised business cycle facts.
Computation of results for New Zealand using UC methodology would be an avenue for further work
and may or may not provide material insights additional to those reported here.
Following application of the HP filter, our detrended series are analysed to obtain their standard
deviation, autocorrelations, and cross-correlations with output. These statistical measures are contained in
Tables 1 for key National Accounts real output and expenditure variables, Table 1a for New Zealand
Treasury fiscal variables, Table 1b for labour market variables, and Table 2 for price and monetary
variables. The standard deviation indicates the volatility of a particular series, while the first-order
autocorrelation tells of its immediate persistence. These measures are supplemented by cross-correlations
which convey the direction and strength of a variable‟s co-movement with output. The correlation
coefficients may be described as either procyclical, countercyclical or acyclical.
3
Because the
contemporaneous correlation is not always the most informative, the co-movement of each series with
real GDP is computed initially as far as fifth-order leads and lags. Specifically, the cyclical component of
the candidate variable at time
time
, for all
is associated with the cyclical component of real GDP at
. Under this approach, a maximum correlation at, for example,
indicates that the cyclical component of the candidate variable tends to lag the aggregate business cycle by
three quarters.
The bivariate cross-correlation coefficient estimates presented in our Tables should be
interpreted in the context of their statistical significance. To this end, the present paper follows the
approach of KBH by reporting for each of the summary statistics Generalised Method of Moments
standard errors, computed as explained in Christiano and Eichenbaum (1992).
A variable will be said to be procyclical when its deviations from trend are contemporaneously correlated with those of
output in a positive fashion; countercyclical when its deviations from trend are contemporaneously correlated with those
of output in a negative fashion; and acyclical when its deviations from trend exhibit a contemporaneous correlation with
output that is close to zero.
3
5
The statistics in the Tables are also limited to the extent that they are full-sample averages. To
assess the stability over time of these sample-average values, we adopt a „moving windows‟ form of
analysis, as utilised in KBH. Specifically, we plot a 21-quarter moving average of the standard deviations
and cross-correlations with output for all variables. Due to the short size of the sample period relative to
KBH, this analysis was also performed for both a 13-quarter and 29-quarter moving average. However, as
the latter results were not materially different, they have not been reported. Further, unlike KBH and the
results reported in most other studies, our stability over time analysis is extended beyond the
contemporaneous correlations to the phase shift corresponding with the maximum cross-correlation
which is statistically significant. For example, in the case of net exports, the moving windows analysis has
been reported for both the contemporaneous correlation and the correlation at a lag of two quarters.
3
Business cycle regularities
Figure 1 shows movements in the logarithm of New Zealand‟s real expenditure-based Gross
Domestic Product (GDP), for the period from 1987q2 to 2010q4. Also plotted is the HP filtered trend
component for
. Over this period, real GDP has grown to 178% of its original level. New
Zealand‟s GDP has been characterised by a number of fluctuations about its long-term trend, the
statistical properties of which can be used to describe, define and analyse that nature of New Zealand‟s
business cycles.
Figure 1
Gross Domestic Product (Expenditure)
1987q2 - 2010q4
10.55
10.45
Natural Logarithm
10.35
10.25
10.15
10.05
GDP
9.95
Trend
9.85
1987:02 1989:02 1991:02 1993:02 1995:02 1997:02 1999:02 2001:02 2003:02 2005:02 2007:02 2009:02
6
The most sizable fluctuations can be reconciled with macroeconomic events of historical
significance4. Such events include the domestically-felt recession experienced simultaneously by the U.S.
and Australia in the early 1990s; the relatively modest slowdown in 1998-1999 associated with the Asian
financial crisis and New Zealand‟s successive summers of drought; the domestic boom of 2007 on the
back of a buoyant housing market and strong consumer confidence; and most recently, the Great
Recession from 2008 through to 2010.
3.1
Volatility and Persistence
Virtually all aspects of the New Zealand economy, real and nominal, have undergone a significant
reduction in volatility since the conclusion of the sample period in KBH (1994). Full sample measures of
the standard deviation and relative standard deviation for New Zealand‟s key macroeconomic variables
are presented in Tables 1, 1a, 1b and 2. The relative standard deviation of macroeconomic variable
is its
standard deviation relative to that of expenditure-based GDP (variable ). These statistical measures
document the emergence of an inherently more stable economy.
Perhaps the most significant stylised fact to have emerged in this context is a marked reduction in
the general volatility of the real economy. As illustrated in Table 1, the standard deviation of the cyclical
component in real GDP has fallen from a remarkable 3.64% in KBH to only 1.41% in the updated
sample period. Unsurprisingly, this fall has coincided with a rise in persistence, with the first-order
autocorrelation of real GDP increasing from .31 in KBH to .77 in the present study. These broad facts
are consistent with the findings of McCaw (2007), and the greater underlying stability of the New Zealand
economy. It should be noted that, with a standard error of 0.10% for the volatility of real GDP, the full
sample average disguises a degree of variability in the standard deviation itself, an observation that is
reinforced by the changeable nature of the moving windows measure illustrated in Figure 2.
The decline in the volatility of real GDP has been mirrored to varying degrees throughout much
of the real economy. Aggregate measures of consumption, investment, and government consumption
expenditure have all witnessed a sizeable reduction in volatility. In light of such conformity, however, two
expenditure components warrant particular attention: private non-residential fixed investment and
government expenditure. Compared to the summary statistics obtained by KBH, the former is the only
variable to experience an increase in its standard deviation, while the latter is the only real variable to
exhibit a fall in its standard deviation relative to that of real output. The disconnectedness of these two
variables echoes remarks made by KBH as to the relative independence of trend volatility in private fixed
investment and government purchases. Such independence may reflect underlying differences in the
determinants of these variables.
4
Descriptive accounts of New Zealand‟s business cycles over the period 1998 to 2011 can be found in Chetwin
(2012), and of past recessions are available in Reddell and Sleeman (2008). Summary statistics for New Zealand‟s
post-war business cycles can be found in Hall and McDermott (2009).
7
What explains the manifest decline in New Zealand‟s real variable volatility? At a proximate level,
this trend may be associated with the stability of prices. As shown in Table 2, the standard deviation of
the GDP deflator has fallen from 5.12% in KBH to a modest 1.20% in the present study.
It may also be associated with improved policy. For example, through the 1989 Reserve Bank Act,
New Zealand became the first country to formally adopt inflation-targeting monetary policy. Two
decades later, it appears the RBNZ has achieved considerable success in achieving its mandate “to
formulate and implement monetary policy directed to the economic objective of achieving and
maintaining stability in the general level of prices” (Reserve Bank of New Zealand Act 1989, s 8). Another
policy-related catalyst for improved stability may be the more prudent use of fiscal policy, following on
from enactment of the Fiscal Responsibility Act 1994, and since incorporated into the Public Finance
Amendment Act 2004. Despite criticism that the rapidity and sequencing of these and other reforms may
have restrained long-term growth over the past 30 years, it appears that the measures have played an
instrumental role in the reducing the volatility of New Zealand‟s business cycles – a feat which, of itself,
may be conducive to growth in the long-run (Ramey & Ramey, 1995). Outside of the policy domain, a
more stable external environment during much of the sample period is likely to have been a further
significant contributing factor.
While measures of absolute volatility can promote an intriguing narrative, business cycle theorists
can often be more interested in measures of relative volatility; such as the standard deviation of the
candidate variable relative to that of real GDP. Of particular note here, is the fact that non-durables
consumption is almost as volatile as real GDP, with a relative standard deviation of 0.95. On the other
hand, consumption of durables is more than twice as volatile as GDP. Consistent with stylised facts for
the majority of industrialised countries, investment is significantly more volatile than both consumption
and real GDP. In fact, with a relative standard deviation of 5.29, investment exhibits substantially more
volatility than in the United States, where it tends to be only about 3 times as volatile as output (King &
Rebelo, 2000).
Results for taxation, transfers, net taxation and government debt to GDP variables were not
reported in either KBH or McCaw. Notable from the measures reported in Table 1a, however, are that all
four variables have been volatile relative to GDP, with net taxation having been particularly so.
The volatility of New Zealand‟s external sector variables has also been notably lower since KBH
(1994), with the exception of the real TWI. Not surprisingly for our sample period of fully flexible
exchange rates, the real TWI has shown somewhat greater volatility, recording 6.06% relative to 5.03%.
From a figure of 3.16%, the standard deviation of the net exports share has fallen to a new sample
average of 1.63%, reflecting a reduction in the volatility of all subcomponents apart from service exports.
Despite this improvement, the standard deviation of New Zealand‟s net export share remains above the
8
median value of 1.06% obtained by Backus et al. (1994) for their sample of 11 OECD countries, and
almost four times as large as the figure of 0.45% for United States.
The decline in net exports volatility was paralleled by an even larger reduction in the standard
deviation of the terms-of-trade from 8.54% to 3.34%. This trend may be ascribed to increased external
stability and some degree of diversification of New Zealand‟s export base. Indeed, upon surveying the
moving windows analysis in KBH, it appears much of the accentuated volatility in that sample came
during the early 1970s when oil prices were subject to severe shocks and New Zealand‟s primary-oriented
export base faced volatile world prices. This trend of high volatility in the terms-of-trade seems to have
re-emerged during the recent global financial crisis, reflected to some extent in Figure 5 by the rise in the
moving standard deviation for that series.
This study also takes the opportunity to revisit the volatility of New Zealand‟s monetary variables
in the wake of the extensive financial market deregulation that took place in the 1980s. KBH document a
substantial degree of volatility in New Zealand‟s monetary sector. The equivalent facts for the period
from 1987q2 to 2010q4 depict a virtually unrecognisable economy in terms of its relative stability. Of
particular note is the significant reduction in the standard deviation of the 90-day nominal interest rate
from 16.2% in KBH to 1.37% in the updated sample period. This was accompanied by a much less
dramatic reduction in the standard deviation of the real 90-day rate from 2.07% to 1.14%. These trends
reflect the increased stability of inflation over the past 20 years, as well as the general trend of declining
volatility in most other macroeconomic variables. Despite the fall in amplitude of both short-term and
long-term interest rates, the short-term rates remain relatively the more volatile, as one would expect.
There has been no material increase in the volatility of labour market variables reported in Table
1b, with all except (FTE) employment having been considerably less volatile than their KBH equivalents.
3.2
Real Variable Regularity
There is nothing particularly striking in the Table 1 cross-correlations between GDP and its
major consumption and investment variables. Both consumption and investment, along with their
disaggregated components, are strongly procyclical, with a magnitude of .76 for aggregate consumption
and .79 for aggregate investment. Similarly, the correlation coefficients for the disaggregated components
range from .58 for „other‟ investment to .83 for durables consumption.5 Of some note, though, is the
magnitude of these coefficients, which evidence a much tighter correlation than was observed by KBH
and bear a closer resemblance to the estimates reported by King and Rebelo (2000) for the United States.
“Real variable regularity” has therefore been both confirmed and enhanced.
„Other‟ investment refers to all Private Gross Fixed Capital Formation that is not classified as either Residential or
Non-Residential.
5
9
3.3
Fiscal Variables
Except in recent multivariate VAR and UC studies for New Zealand by Claus et al. (2006), Dungey and
Fry (2009) and Hall and McDermott (2011), it has been common to investigate stylised facts solely for
National Accounts government expenditure variables. However, even for these variables the stylised facts
continue to be controversial. For example, have government expenditure variables been consistently pro-,
counter- or a-cyclical, have there been robust lead or lag lengths, and what has been the degree of
movement in these relationships over time? With the aim of shedding further light on each of these
aspects, we first present results for the conventionally reported government expenditure variables, and
then for the other fiscal variables published in HMcD (2011, Table 7).
3.3.1
Government Expenditure
In KBH, government expenditure variables displayed no systematic cyclical tendency (0.16
correlation for the aggregate and 0.14 for central government). The maximum statistically significant
cross-correlation was 0.39 with fluctuations in government expenditure lagging fluctuations in real GDP
by five quarters.
In this sense, government expenditures constituted an exception to the
contemporaneous real variable regularities displayed by consumption, investment, exports, and imports.
That acyclical behaviour of government spending was consistent with the finding of Backus and Kehoe
(1992) that internationally, government purchases exhibit no systematic cyclical tendency.
The results we present for government expenditure variables in Tables 1 and 1a confirm both
that aggregate government expenditure has been acyclical and that the maximum cross-correlation has
been positive with a lag of five quarters. The magnitude of that maximum has, however, increased from
the previous 0.39 to around 0.56, and the latter magnitude should be seen in the context of its variability
over time, as illustrated in Figures 4 and 4a. There, the 21-quarter moving average of the five-quarter
lagged cross-correlation for aggregate government expenditure has fluctuated considerably, and has
almost always exhibited quite pronounced positive co-movements. The key changes in direction of the
aggregate government expenditure variable seem associated most strongly with those in its investment
expenditure component.
What then might this relationship imply for the nature of fiscal policy in New Zealand? Subject
of course to the bivariate nature of the measure, and having due regard to the interpretation of most
recent movements being limited by end-point issues associated with moving windows methodology6, the
correlation is consistent with governments‟ following above trend movements in real GDP with often
quite substantial above trend government expenditures.
6
Further work is planned to address end-point issues.
10
3.3.2
Other Fiscal Variables
The net taxation variable has been consistently and significantly procyclical, centred around its
sample average correlation of 0.57 and with considerable variability over time (Table 1a and Figure 3a).
Movements in taxation seem to have dominated those in transfer payments, as movements in taxation
have been similarly positively correlated with a lag of one quarter (0.57). The maximum cross-correlation
for transfers has been negative, with a lead of five quarters (-0.41)7.
Movements in the gross government debt to GDP ratio have also been very variable over time,
and have been primarily negatively correlated with a lag of three quarters (-0.60), i.e. above trend
movements in real GDP have been associated most strongly with subsequent below trend movements in
the government debt to GDP ratio (Table 1a and Figure 4a) .
3.4
Open Economy Variables
Among the industrialised countries of the world, the business cycle literature contains systematic
evidence of a countercyclical trade balance. This stylised fact was reported by Backus and Kehoe (1992)
for 10 industrialised countries. Nowadays, the assumption of a countercyclical trade balance is a
commonplace feature of many DSGE models.
In light of this empirical consensus, one of the more puzzling stylised facts to have emerged from
KBH was the sample average acyclical nature of net exports (-0.09, not significant). This sample average
value reflected substantial subperiods of both pro-and counter-cyclicality, ending with a period from
(centred) 1983 onwards of strong procyclicality (.60). These observations were inconsistent with Backus,
Kehoe, and Kydland‟s (1992) 11-country median contemporaneous correlation of -0.29, and provide a
key justifications for this paper‟s reassessment of New Zealand‟s business cycle facts.
This paper finds a more robust countercyclical tendency in net exports for the period from
1987q2 to 2010q4. The contemporaneous cross-correlation of -.33 in Table 1 reveals a weakly
countercyclical, yet statistically significant, relationship between the net exports share and real GDP.
Despite a notable degree of variability in the moving windows of this correlation, it remains negative for
almost the entire sample period. On this evidence, the behaviour of net exports more closely reflects the
common experience of other OECD countries in the updated sample period.
In a proximate sense, the stronger counter-cyclicality of New Zealand‟s trade balance, relative to
that found in KBH, can be attributed primarily to the relatively stronger procyclicality of imports. As can
be seen from the relevant panels in Figure 3, imports determine not only the sign, but also the strength of
the net exports share. In periods where imports and real GDP are closely correlated, above average
7
This result for transfers is in contrast to the statistically significant result reported in HMcD (2011, Table 7) of
transfers being weakly contemporaneously countercyclicaal (-0.16). The HMcD sample period was 1983q2 to
2006q4.
11
output is more likely to result in a deterioration of the trade balance. The pivotal role of imports in the
determination of the relationship between net exports and output can be reconciled with the strong
income term assigned to import demand by empirical economists such as Krugman, Baldwin, Bosworth,
and Hooper (1987).
But is the contemporaneous relation the strongest one? As in KBH, where net exports are
negatively cross-correlated with a magnitude of -0.35 and a lag of two quarters, we report a markedly
stronger maximum cross-correlation of -.52, also at a lag of two quarters. The six-month lag might in
principle be explainable in terms of concepts underpinning a J-curve. For example, if the terms-of-trade
are procyclical, then for an expansion phase of the aggregate business cycle net exports could fall below
trend as quantities of imported goods also expand.8 In New Zealand, however, a deterioration of the
trade balance could be dampened in the short-run by the relative stickiness of trade quantities. When
trade quantities are slow to adjust, the price effect of a rise in the terms-of-trade would operate to
improve the trade balance, thereby partially offsetting the adverse quantity effect outlined above. This
argument would not be inconsistent with a two-quarter lag in the negative co-movement of the trade
balance. So, has New Zealand‟s terms of trade been procyclical or leading with a positive crosscorrelation? The answer would be yes if one turned to the maximum cross-correlation for the terms of
trade reported in HMcD (2011, Table 7) for the sample period 1983q2 to 2006q4. There, the terms of
trade was statistically significant with value 0.33, and led by two quarters. However, as reported in our
Table 2, this relationship is not sustained when observations are extended through to 20010q4. There is
then no statistically significant contemporaneous relation, and the maximum cross-correlation of 0.28 for
a lag of two or three quarters is puzzling.
A robust specific role for the terms of trade has clearly not yet emerged from our bivariate work.
We have, however, not yet assessed the relationship between the cyclical component of net exports at
time
and the cyclical component of the terms-of-trade at time , as encouraged by Backus et al.
(1994), and some form of multivariate approach may be even more fruitful..
We have also considered the cyclical behaviour of both the real and nominal TWI. Both of these
measures have been quite strongly procyclical (Table 2, 0.56), with magnitudes varying quite considerably
over time (Figure 6). Not surprisingly, this is a considerably stronger result than the generally nonsignificant negatively correlated results obtained by KBH for their sample period which covered a series
of different exchange rate regimes.
Disaggregated components of the trade balance also warrant brief discussion. Imports of services
co-move positively with a two-period lag, contrasting with imports of goods which move procyclically.
SNZ calculates the terms-of-trade as the price of exports relative to the price of imports. Accordingly, a rise in the
terms-of-trade constitutes an improvement while a fall constitutes a deterioration. This explains why the
abovementioned argument hinges upon a procyclical terms-of-trade as opposed to a countercyclical measure.
8
12
More interestingly, though, exports of services have a maximum cross-correlation of 0.48 at a lead of two
quarters, and are the only major real output variable in our set to lead the aggregate cycle.
3.5
Monetary Variables
The study of KBH was unable to produce robust monetary variable results, due to major
financial market deregulation during the sample period and changes to New Zealand‟s exchange rate
regime. Our using a sample period from 1987q2 to 2010q4 enables us to reassess monetary sector stylised
facts in the context of a more stable regulatory environment and a floating exchange rate regime.
As shown in Table 2, all interest rate variables are contemporaneously procyclical with output.
However, the contemporaneous relationship between interest rates and output is not the most striking.
Consistent with the findings of McCaw (2007), our interest rates tend to lag economic activity by
approximately three quarters. For a phase shift of this nature, the correlation coefficients range from 0.57
for the nominal rate on five-year government bonds to 0.73 for the real 90-day rate. On the basis of these
results, one might be tempted to infer that the response of monetary policy to detrended output tends to
be reactive, rather than proactive9. However caution must be exercised when associating changes in the
interest rate with changes in monetary policy; as it is entirely possible for the former to occur
independently of the latter via changes in credit demand or other financial market factors. It may also
mean, for a flexible inflation-targeting central bank, that responses of output to interest rate movements
may have been quite weak over the short- to medium-term.
To assess further this three-quarter lagged correlation, consider the representative case of the real
90-day interest rate. As illustrated in Figure 7, this has shown considerable stability over time, with the
exception of a centred period between 2001q3 and 2005q2 during which it became close to acyclical.
Nonetheless, one may posit an explanation grounded upon RBNZ‟s considerable focus on the housing
boom during this period. In the context of an already tight domestic economy, the housing boom
accentuated inflation but added little to the economy in terms of increased output . Accordingly, the
RBNZ‟s interest rate response may have been less dependent on the deviations in output from trend, and
more dependent upon the deviations from trend inflation.
There are several other distinctive observations for the cyclicality of monetary variables. The real
interest rate displays a slightly stronger procyclical correlation than its nominal counterpart, short and
long-term rates have broadly similar degrees of procyclicality, and results for the yield gap variable are
consistent with those for the other interest rate variables. Finally, the two real monetary aggregate
9
HMcD (2011, Table 7) report, for the period 1983q2 – 2006q4, a maximum statistically significant crosscorrelation of -0.31 for a lead of nine quarters, which is consistent with weakly proactive monetary policy. However,
in HMcD‟s multivariate UC results, this monetary policy variable was then not significant, and dominated by terms
of trade, government expenditure and net migration variables.
13
variables included in our data set – M3(r) and DC(r) – are basically acyclical, but their correlation with
output climbs to 0.61 and 0.40 respectively at a lag of five quarters.
3.6
Prices
Cross correlations between fluctuations in aggregate price level and output variables can be
consistent with a range of theoretical underpinnings and empirical interpretations. KBH report a relatively
weak, but statistically significant countercyclical relationship between these two variables, but the moving
average relationships showed considerable variability over time. For example, the CPI exhibited a period
of sustained procyclicality for the centred period covering the late 1970s and early 1980s.
Three price level measures are used in this study: the All Groups Consumer Price Index (CPI);
the CPI excluding food, fuel, and government charges; and the GDP deflator. Their respective
contemporaneous cross-correlations of 0.04, -0.14 and .12 are not statistically significant and are
consistent with fluctuations in the price level being acyclical. However, when one considers the maximum
co-movement at a lag of five quarters, the resulting coefficients offer evidence of a modest positive
correlation with output for all three measures of prices (i.e., from 0.38 to 0.45)10. For example, when
output is above trend at time , one would expect the CPI to be somewhat above trend at time
a
lagged relationship which is consistent with a degree of price stickiness at aggregate levels. Again not
surprisingly, however, the lagged correlation has varied considerably over time, and has remained positive
except for a short period centred in the early 2000s.
3.7
Labour Market Variables
The greater part of our sample period post-dates passage of New Zealand‟s Employment
Contracts Act and the subsequent Employment Relations Act, whereas the KBH sample pre-dated those
Acts. So could one have expected materially different results for our contemporary sample period?
Key findings from KBH were that: aggregate employment fluctuations were positively correlated
with output fluctuations with a lag of one quarter (0.40); aggregate labour productivity was strongly
procyclical (0.79); and aggregate real average hourly compensation was very weakly procyclical (0.15
(0.10)), reflecting a lengthy period of procylicality and two short periods of centred countercyclicality.
From a producer perspective, real private unit labour costs were consistently contemporaneously
countercyclical for almost all of the period (-0.42), with this average having reflected contemporaneously
procyclical real private labour costs (0.34) and their movements over time having been almost a mirror
image of labour productivity movements.
Some interesting results emerge from our focus on the post-1987 sample period. Perhaps least
surprising is that (FTE) employment fluctuations have on average now been more strongly positively
10
There is somewhat less strong but statistically significant evidence that the aggregate price level is negatively crosscorrelated with a lead of approximately five quarters (e.g. -0.35 (0.09) for the CPI).
14
correlated with output fluctuations, but still with a lag of two quarters (0.59 relative to the KBH 0.40).
Average procyclical (FTE) labour productivity is now shown to be less strongly correlated than the
average (hourly) labour productivity reported in KBH (0.54 relative to 0.79), and is further reflected in
more recent real unit labour costs having been more strongly countercyclical (-0.68 versus
-0.42). But perhaps the most materially different results, which are also consistent with results presented
in McCaw (2007), are those for our two real wage variables: firstly, real average hourly earnings are no
longer weakly procyclical but significantly negatively correlated with output movements at a lag of two
quarters (-0.32, Table 1b); and secondly, the KBH (private) real labour costs relationship of modest
procyclicality (0.34) is in contrast to our post-ECA real labour costs variable being modestly negatively
correlated with a lag of four quarters (-0.40, Table 1b).
4
Conclusion
Drawing on well-established univariate and bivariate growth cycle methodology, we have
established a set of stylised facts for New Zealand business cycles for the period from 1987q2 to 2010q4.
Our results, taken in conjunction with recent findings from McCaw (2007) and Hall and McDermott
(2011), are benchmarked against those established by Kim, Buckle and Hall (1994) for the period 1966q4
to 1990q1, and suggest a number of new findings. Some of these findings challenge prior understandings,
while others clarify previous uncertainties.
By way of broad perspective, our results show that for a sample period following the major
economic reforms of the mid-1980s and early-1990s, fluctuations in New Zealand‟s aggregate business
cycle have exhibited greater persistence than previously. Not surprisingly, this has been associated with
reduced volatility for key macroeconomic variables, particularly so for government expenditure, nominal
interest rates, net exports, terms of trade, labour productivity, real wage and other price variables. Net
taxation, government investment expenditure, and the components of private gross fixed capital
formation have been volatile, relative to GDP.
Government sector, open economy, monetary and labour market cross correlation results differ
in important respects from those established previously. In particular, monetary sector stylised facts are
somewhat more robust. Information from non-contemporaneous cross-correlations is shown to be
additionally important for some key variables, as is information from movements in cross-correlations
over time.
More specifically, real variable regularity has been confirmed, and the procyclical co-movements
for private consumption and investment have been stronger. Government expenditure has remained
(contemporaneously) acyclical, and it continues to be the case that government expenditure has co-moved
positively with a lag of five quarters. Fluctuations in real net taxation have been consistently
contemporaneously procyclical, with considerable variability over time. Movements in the gross
15
government debt to GDP ratio have also been very variable over time, and primarily negatively correlated
with a lag of three quarters. Imports have now become convincingly more procyclical than exports, and
as a result the net exports share has become both more strongly countercyclical and more strongly
negatively cross correlated with a lag of two quarters. It has now also been established that real TWI
movements have been quite strongly procyclical, with considerable variability over time. Interest rate
variables have for the most part been weakly procyclical, but they have also exhibited a considerably
stronger positive correlation with output at a lag of three quarters. These monetary sector regularities are
of material significance, given the uncertainty that characterised the monetary sector during the KBH
sample period. Finally, results for our labour market variables are preliminary in nature but suggest the
possibility of considerably different stylised facts for labour productivity, real labour costs, real unit labour
costs and real average hourly earnings.
Our univariate and bivariate stylised facts should not, of course, be used on their own as
benchmarks for macroeconomic models. They can be further assessed against key findings from
multivariate structural and reduced form approaches, with due regard for the fact that bivariate
correlations provide statistical associations but not necessarily causative relations. The robustness of our
underlying deviations from trend series could also be further verified, using an unobserved components
approach to detrending.
Overall, however, a more credible set of benchmark regularities have been established, to help
underpin the construction and use of contemporary New Zealand macroeconomic models.
16
A
Tables
TABLE 1
Cyclical Behaviour of Key Output and Expenditure Variables
Quarterly Deviations from Trend: 1987(2)-2010(4), λ=1600
Volatility Relative Volatility
Variable x
Auto-
SD% SD(x)%/SD(y)% correlation
GDP (production)
1.44%
1.02
0.86
GDP (expenditure)
1.41%
1.00
0.77
1.03
0.81
Cross-correlation of Real GDP (expenditure) with
x t-5
x t-4
x t-3
x t-2
x t-1
xt
x t+1
x t+2
x t+3
x t+4
x t+5
0.03
0.10
0.31
0.49
0.62
0.76
0.71
0.65
0.58
0.49
0.36
(0.10%)
Consumption
1.45%
(0.09%)
Non-durables
1.34%
(0.09) (0.09) (0.09) (0.09) (0.10) (0.04) (0.06) (0.08) (0.08) (0.09) (0.09)
0.95
0.64
(0.09%)
Durables
3.28%
2.33
0.77
(0.29%)
Private gross fixed capital formation
7.46%
8.59%
5.31
0.76
6.11
0.76
11.14%
9.00%
7.92
0.74
2.64%
6.40
0.57
1.88
0.54
1.37%
11.01%
0.97
0.43
1.63%
7.83
0.62
2.49%
1.16
0.67
1.77
0.46
3.05%
4.61%
2.17
0.33
4.88%
3.28
0.76
3.47
0.71
5.76%
4.56%
0.44
0.56
0.69
0.83
0.72
0.60
0.50
0.34
0.22
0.14
0.26
0.37
0.48
0.65
0.79
0.69
0.61
0.45
0.26
0.09
0.00
0.11
0.28
0.47
0.62
0.74
0.66
0.53
0.29
0.13 -0.01
0.22
0.31
0.39
0.49
0.53
0.62
0.54
0.56
0.55
0.45
0.36
0.16
0.25
0.28
0.28
0.45
0.58
0.48
0.43
0.31
0.14 -0.01
-0.12 -0.15 -0.12 -0.01
0.05
0.14
0.15
0.24
0.28
0.34
0.54
-0.17 -0.18 -0.22 -0.13 -0.02
0.07
0.08
0.27
0.28
0.32
0.50
-0.13 -0.15 -0.08
0.04
0.09
0.18
0.21
0.23
0.29
0.33
0.48
0.10
0.02 -0.08 -0.14 -0.32 -0.33 -0.43 -0.52 -0.45 -0.34 -0.26
0.05
0.04
0.04
0.15
0.19
0.31
0.13 -0.04 -0.10 -0.16 -0.20
-0.10 -0.15 -0.17 -0.06
0.03
0.20
0.05 -0.09 -0.10 -0.08 -0.06
0.39
0.47
0.47
0.48
0.43
0.33
0.27
0.13
0.00 -0.13 -0.29
-0.05
0.05
0.15
0.27
0.47
0.52
0.54
0.56
0.44
0.30
0.21
(0.11) (0.11) (0.10) (0.08) (0.07) (0.06) (0.08) (0.08) (0.09) (0.09) (0.09)
4.10
0.66
(0.37%)
Imports of Services
0.26
(0.09) (0.08) (0.09) (0.08) (0.09) (0.10) (0.10) (0.10) (0.10) (0.10) (0.10)
(0.32%)
Imports of Goods
0.09
(0.11) (0.11) (0.10) (0.11) (0.12) (0.10) (0.13) (0.12) (0.10) (0.11) (0.09)
(0.36%)
Imports of Goods and Services
0.39
(0.12) (0.11) (0.10) (0.11) (0.11) (0.09) (0.12) (0.11) (0.10) (0.10) (0.09)
(0.32%)
Exports of Services
0.43
(0.11) (0.12) (0.10) (0.09) (0.08) (0.07) (0.08) (0.08) (0.09) (0.09) (0.09)
(0.23%)
Exports of Goods
0.48
(0.10) (0.10) (0.10) (0.10) (0.10) (0.10) (0.11) (0.12) (0.11) (0.10) (0.09)
(0.13%)
Exports of Goods and Services
0.51
(0.11) (0.12) (0.10) (0.09) (0.10) (0.09) (0.10) (0.10) (0.09) (0.09) (0.08)
(0.90%)
Net Exports Share
0.55
(0.10) (0.11) (0.11) (0.09) (0.10) (0.09) (0.10) (0.10) (0.09) (0.10) (0.09)
(0.13%)
Govt. Investment
0.60
(0.14) (0.10) (0.10) (0.10) (0.08) (0.06) (0.08) (0.11) (0.09) (0.09) (0.09)
(0.23%)
Govt. Consumption
0.53
(0.13) (0.13) (0.12) (0.10) (0.09) (0.07) (0.08) (0.08) (0.10) (0.09) (0.10)
(0.67%)
Government Expenditure
0.41
(0.11) (0.11) (0.09) (0.08) (0.06) (0.03) (0.06) (0.07) (0.08) (0.09) (0.10)
(0.84%)
Other
0.18
(0.13) (0.12) (0.10) (0.09) (0.08) (0.03) (0.07) (0.10) (0.09) (0.09) (0.10)
(0.56%)
Non-residential
0.02
(0.10) (0.09) (0.09) (0.09) (0.11) (0.03) (0.06) (0.09) (0.09) (0.09) (0.09)
(0.48%)
Residential
-0.03
(0.10) (0.09) (0.09) (0.09) (0.08) (0.06) (0.08) (0.09) (0.09) (0.09) (0.10)
-0.04
0.07
0.17
0.25
0.47
0.51
0.51
0.51
0.39
0.24
0.18
(0.12) (0.12) (0.10) (0.09) (0.08) (0.06) (0.08) (0.09) (0.09) (0.09) (0.09)
3.24
0.66
(0.33%)
-0.04 -0.03
0.05
0.21
0.30
0.35
0.42
0.47
0.38
0.36
0.22
(0.09) (0.09) (0.08) (0.08) (0.08) (0.07) (0.08) (0.09) (0.09) (0.09) (0.10)
Notes: Numbers in parentheses are GMM standard errors; column 3 is a first order serial correlation
17
TABLE 1a
Cyclical Behaviour of Key New Zealand Treasury Fiscal Variables
Quarterly Deviations from Trend: 1987(2)-2010(4), λ=1600
Volatility
Variable x
GDP (expenditure)
Relative Volatility
Auto-
SD% SD(x)%/SD(y)% correlation
1.41%
1.00
0.77
1.74
0.51
Cross-correlation of Real GDP (expenditure) with
x t-5
x t-4
x t-3
x t-2
x t-1
xt
x t+1
x t+2
x t+3
x t+4
x t+5
-0.20 -0.23 -0.21 -0.09
0.01
0.13
0.20
0.32
0.35
0.37
0.57
(0.10%)
Government Expenditure
2.45%
(0.25%)
Govt. Consumption
1.51%
(0.10) (0.11) (0.11) (0.10) (0.11) (0.09) (0.09) (0.10) (0.09) (0.11) (0.09)
1.07
0.37
(0.15%)
Govt. Investment
11.60%
8.25
0.56
(1.01%)
Net Govt. Expenditure/GDP
1.28%
(Govt. expend. - net tax)
(0.10%)
-0.16 -0.17 -0.21 -0.15 -0.04
0.03
0.02
0.22
0.22
0.25
0.44
(0.11) (0.12) (0.10) (0.09) (0.10) (0.09) (0.10) (0.10) (0.09) (0.09) (0.09)
-0.24 -0.25 -0.17 -0.04
0.05
0.19
0.31
0.35
0.39
0.39
0.50
(0.09) (0.10) (0.10) (0.10) (0.11) (0.10) (0.10) (0.11) (0.11) (0.11) (0.10)
0.40
-0.15 -0.30 -0.35 -0.45 -0.52 -0.51 -0.48 -0.40 -0.27 -0.13
0.04
(0.10) (0.10) (0.10) (0.09) (0.08) (0.08) (0.09) (0.09) (0.10) (0.11) (0.11)
Government Revenue variables
Taxation
3.92%
2.79
0.41
2.39
0.50
(0.31%)
Transfers
3.36%
(Taxation - transfers)
Gross Govt. Debt/GDP
6.58%
0.18
0.36
0.49
0.55
0.57
0.53
0.41
0.31
0.20
-0.41 -0.38 -0.31 -0.21 -0.15 -0.12 -0.05 -0.02 -0.02
0.07
0.09
(0.08) (0.09) (0.11) (0.11) (0.10) (0.09) (0.09) (0.09) (0.09) (0.10) (0.09)
4.68
0.46
(0.48%)
2.99%
0.10
(0.10) (0.11) (0.11) (0.10) (0.08) (0.07) (0.08) (0.09) (0.10) (0.12) (0.10)
(0.37%)
Net taxation
-0.04
0.08
0.22
0.28
0.42
0.53
0.57
0.56
0.51
0.39
0.26
0.16
(0.11) (0.11) (0.11) (0.10) (0.08) (0.07) (0.08) (0.08) (0.09) (0.11) (0.11)
2.13
0.79
(0.28%)
0.26
0.10 -0.05 -0.15 -0.34 -0.45 -0.51 -0.54 -0.60 -0.50 -0.43
(0.10) (0.11) (0.11) (0.11) (0.09) (0.07) (0.08) (0.10) (0.08) (0.08) (0.08)
Notes: Numbers in parentheses are GMM standard errors; column 3 is a first order serial correlation
18
TABLE 1b
Cyclical Behaviour of Key New Zealand Labour Market Variables
Quarterly Deviations from Trend: 1987(2)-2010(4), λ =1600
Volatility
Variable x
GDP (expenditure)
Relative Volatility
Auto-
SD% SD(x)%/SD(y)% correlation
1.41%
1.00
0.77
0.90
0.88
Cross-correlation of Real GDP (expenditure) with
x t-5
x t-4
x t-3
x t-2
x t-1
xt
x t+1
x t+2
x t+3
x t+4
x t+5
0.18
0.28
0.35
0.45
0.53
0.57
0.58
0.59
0.56
0.50
0.42
(0.10%)
Employment
1.27%
(0.08%)
Unemployment
0.70%
(0.11) (0.11) (0.10) (0.08) (0.07) (0.07) (0.08) (0.09) (0.09) (0.09) (0.09)
0.50
0.88
(0.05%)
Labour Productivity
1.08%
0.77
0.65
(0.08%)
Real Average Weekly Earnings †
0.88%
0.88%
0.63
0.41
0.63
0.65
0.99%
1.76%
0.70
0.62
1.25
0.77
0.39
0.54
0.36
0.18
0.02 -0.12 -0.25
-0.08 -0.04
0.01
0.00 -0.03 -0.10 -0.26 -0.32 -0.27 -0.22 -0.20
-0.24 -0.26 -0.23 -0.27 -0.29 -0.32 -0.39 -0.45 -0.38 -0.25 -0.14
0.03
0.00 -0.01
0.00
0.00 -0.09 -0.12 -0.27 -0.37 -0.40 -0.37
(0.10) (0.10) (0.09) (0.09) (0.10) (0.10) (0.11) (0.11) (0.11) (0.12) (0.11)
(0.12%)
0.07
0.00 -0.13 -0.31 -0.48 -0.68 -0.59 -0.56
0.50 -0.40
0.25
(0.10) (0.10) (0.09) (0.08) (0.07) (0.05) (0.09) (0.11) (0.11) (0.11) (0.10)
Notes: Numbers in parentheses are GMM standard errors; column 3 is a first order serial correlation
†
Sample period 1989(1)-2010(4)
††
0.24
(0.09) (0.09) (0.10) (0.10) (0.09) (0.09) (0.08) (0.07) (0.09) (0.09) (0.10)
(0.07%)
Real Unit Labour Cost††
0.12
(0.11) (0.11) (0.11) (0.11) (0.10) (0.11) (0.10) (0.09) (0.10) (0.11) (0.11)
(0.06%)
Real Labour Cost††
-0.01 -0.01
(0.10) (0.09) (0.09) (0.09) (0.10) (0.07) (0.08) (0.09) (0.10) (0.10) (0.12)
(0.06%)
Real Average Hourly Earnings †
-0.24 -0.35 -0.42 -0.53 -0.62 -0.67 -0.68 -0.62 -0.54 -0.43 -0.32
(0.13) (0.12) (0.11) (0.08) (0.06) (0.05) (0.07) (0.09) (0.10) (0.10) (0.10)
Sample period 1992(4)-2010(4)
19
Table 2
Cyclicality of Key Price and Monetary Variables
Quarterly Deviations from Trend: 1987(2)-2010(4), λ =1600
X12 Seasonally Adjusted
Variable x
Volatility
Auto-
SD%
correlation
Cross-correlation of Real GDP (expenditure) with
x t-5
x t-4
x t-3
x t-2
x t-1
xt
x t+1
x t+2
x t+3
x t+4
x t+5
Price Variables
3.34%
TOT
0.73
(0.29%)
6.05%
TWI
6.06%
1.26%
0.83
0.68%
0.70
1.20%
0.13
0.14
0.20
0.28
0.28
0.20
0.12
-0.14
0.00
0.20
0.38
0.50
0.56
0.56
0.51
0.39
0.28
0.21
-0.14
0.01
0.22
0.39
0.51
0.56
0.57
0.53
0.41
0.30
0.23
-0.35
-0.29
-0.27
-0.23
-0.10
0.04
0.15
0.25
0.34
0.40
0.43
(0.09) (0.09) (0.08) (0.08) (0.08) (0.08) (0.07) (0.08) (0.09) (0.10) (0.11)
0.70
(0.08%)
GDP deflator
0.06
(0.08) (0.08) (0.08) (0.07) (0.07) (0.05) (0.06) (0.08) (0.08) (0.09) (0.11)
(0.21%)
CPI less food, fuel, govt charges*
0.05
(0.08) (0.08) (0.08) (0.07) (0.06) (0.05) (0.06) (0.07) (0.08) (0.09) (0.11)
(0.45%)
CPI
0.03
(0.10) (0.10) (0.10) (0.09) (0.08) (0.09) (0.10) (0.10) (0.09) (0.09) (0.07)
0.84
(0.48%)
Real TWI
-0.02
-0.22
-0.24
-0.22
-0.11
-0.11
-0.14
0.01
0.15
0.23
0.34
0.38
(0.10) (0.11) (0.09) (0.10) (0.11) (0.11) (0.10) (0.08) (0.08) (0.09) (0.10)
0.69
(0.09%)
-0.23
-0.15
-0.05
-0.02
0.02
0.12
0.13
0.27
0.39
0.41
0.45
(0.09) (0.10) (0.09) (0.09) (0.10) (0.09) (0.09) (0.09) (0.08) (0.09) (0.09)
Monetary Variables
90-day Bank Bill
1.37%
0.79
(0.09%)
5 yr Govt Bond
0.77%
0.77
(0.06%)
Floating first mortgage rate
1.12%
0.88%
1.14%
0.66
2.38%
0.83
2.09%
(0.15%)
0.20
0.37
0.51
0.61
0.68
0.61
0.48
-0.23
-0.17
-0.07
0.10
0.29
0.44
0.50
0.54
0.57
0.49
0.35
-0.24
-0.21
-0.16
-0.07
0.09
0.28
0.46
0.58
0.69
0.07
0.56
-0.08
-0.05
0.00
0.00
0.06
0.18
0.35
0.47
0.56
0.52
0.43
-0.24
-0.15
-0.05
0.07
0.23
0.39
0.54
0.66
0.73
0.66
0.53
(0.10) (0.10) (0.11) (0.11) (0.11) (0.09) (0.07) (0.07) (0.09) (0.10) (0.11)
0.82
(0.14%)
DC(r)***
0.06
(0.09) (0.09) (0.10) (0.11) (0.10) (0.10) (0.09) (0.09) (0.09) (0.09) (0.09)
(0.07%)
M3(r)***
-0.04
(0.09) (0.10) (0.10) (0.11) (0.11) (0.10) (0.08) (0.08) (0.08) (0.10) (0.11)
(0.07%)
Real 90-day Bank Bill**
-0.13
(0.10) (0.11) (0.11) (0.11) (0.11) (0.09) (0.07) (0.08) (0.09) (0.11) (0.13)
0.84
(0.06%)
90-day yield gap
-0.19
(0.10) (0.10) (0.11) (0.11) (0.11) (0.10) (0.08) (0.08) (0.09) (0.10) (0.11)
-0.24
-0.20
-0.21
-0.09
0.03
0.19
0.29
0.38
0.45
0.55
0.61
(0.08) (0.09) (0.10) (0.10) (0.10) (0.10) (0.09) (0.09) (0.09) (0.09) (0.09)
0.82
-0.25
-0.20
-0.20
-0.13
-0.02
0.05
0.13
0.19
0.28
0.37
0.40
(0.07) (0.07) (0.08) (0.08) (0.08) (0.09) (0.09) (0.09) (0.10) (0.10) (0.10)
Notes: Numbers in parentheses are GMM standard errors; Column 2 is a first-order Autocorrelation
* Observations available from 1989(1) to 2010(4) only
** Observations available from 1987(3) to 2010(4) only
*** Observations available from 1988(2) to 2010(4) only
20
B
Figures
Figure 2
Moving Percentage Standard Deviations from Trend: Real Variables
Real GDP (Expenditure)
0.021
0.019
0.017
0.015
0.013
0.011
0.009
0.007
0.005
1989:04
Real Private Consumption
0.019
0.017
0.015
0.013
0.011
0.009
0.007
1994:02
1998:04
2003:02
2007:04
0.005
1989:04
1994:02
1998:04
2003:02
2007:04
Private Gross Fixed Capital Formation
0.12
0.1
0.08
0.06
0.04
0.02
0
1989:04
1994:02
1998:04
2003:02
2007:04
Net Exports Share
0.023
0.021
0.019
0.017
0.015
0.013
0.011
0.009
0.007
0.005
1989:04
Total Exports
0.04
0.035
0.03
0.025
0.02
1994:02
1998:04
2003:02
0.015
1989:04
2007:04
1994:02
Total Imports
0.07
0.065
0.06
0.055
0.05
0.045
0.04
0.035
0.03
0.025
1989:04
1994:02
1998:04
21
2003:02
2007:04
1998:04
2003:02
2007:04
Figure 2a
Moving Percentage Standard Deviations from Trend: New Zealand Treasury Fiscal Variables
Government Consumption
Government Investment
0.025
0.02
0.015
0.01
0.005
0
1989:04
1994:02
1998:04
2003:02
2007:04
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
1989:04
Real Government Consumption and
Investment
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
1989:04
1994:02
1998:04
2003:02
2007:04
Real Taxation
0.06
0.05
0.04
0.03
0.02
1994:02
1998:04
2003:02
0.01
1989:04
2007:04
Real Transfers
1994:02
1998:04
2003:02
2007:04
Real Net Taxation
0.09
0.06
0.08
0.05
0.07
0.04
0.06
0.03
0.05
0.04
0.02
0.03
0.01
1989:04
1994:02
1998:04
2003:02
0.02
1989:04
2007:04
1994:02
Real Government Consumption and
Investment less Real Net Taxation
2003:02
2007:04
Debt
0.06
0.02
0.05
0.015
0.04
0.03
0.01
0.02
0.005
0
1989:04
1998:04
0.01
1994:02
1998:04
2003:02
0
1989:04
2007:04
22
1994:02
1998:04
2003:02
2007:04
Figure 2b
Moving Percentage Standard Deviations from Trend: Labour Market Variables
Employment
Unemployment
0.025
1.4
1.2
0.02
1
0.015
0.8
0.01
0.6
0.4
0.005
0
1989:04
0.2
1994:02
1998:04
2003:02
0
1989:04
2007:04
1994:02
1998:04
2003:02
2007:04
Labour Productivity
0.016
0.014
0.012
0.01
0.008
0.006
0.004
0.002
0
1989:04
1994:02
1998:04
2003:02
2007:04
Real Average Weekly Earnings
Real Average Hourly Earnings
0.012
0.014
0.01
0.012
0.01
0.008
0.008
0.006
0.006
0.004
0.004
0.002
0.002
0
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
0
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
Real Labour Cost
Real Unit Labour Cost
0.014
0.025
0.012
0.02
0.01
0.008
0.015
0.006
0.01
0.004
0.005
0.002
0
1995:02
1997:04
2000:02
2002:04
2005:02
0
1995:02
2007:04
23
1997:04
2000:02
2002:04
2005:02
2007:04
Figure 3
Moving Contemporaneous Cross Correlations with Real GDP: Real Variables
Private Gross Fixed Capital Formation
Real Private Consumption
1
1
0.9
0.8
0.8
0.7
0.6
0.6
0.4
0.2
1989:04
0.5
1994:02
1998:04
2003:02
2007:04
0.4
1989:04
0.6
0.4
0.2
0
-0.2
-0.4
1994:02
1998:04
2003:02
2007:04
0.5
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
-0.5
1989:04
0.8
0.1
0.6
0
0.4
-0.1
0.2
-0.2
0
-0.3
-0.2
-0.4
-0.4
-0.5
1994:02
1998:04
2003:02
1994:02
2007:04
-0.6
1989:04
1994:02
2007:04
1998:04
2003:02
2007:04
1998:04
2003:02
2007:04
Total Imports
Total Exports
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0
-0.2
-0.4
1989:04
2003:02
Net Exports Share
Government Investment
-0.6
1989:04
1998:04
Government Consumption
Government Expenditure
-0.6
1989:04
1994:02
1994:02
1998:04
2003:02
2007:04
-0.2
1989:04
24
1994:02
1998:04
2003:02
2007:04
Figure 3a
Moving Contemporaneous Cross Correlations with Real GDP: New Zealand Treasury Fiscal Variables
Government Consumption
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
1989:04
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
1989:04
1994:02
1998:04
2003:02
Government Investment
2007:04
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
1989:04
Real Government Consumption and
Investment
1994:02
2003:02
2007:04
Real Taxation
0.9
0.8
0.7
0.6
0.5
0.4
0.3
1994:02
1998:04
2003:02
0.2
1989:04
2007:04
1994:02
Real Transfers
0.2
0
-0.2
-0.4
-0.6
1994:02
1998:04
1998:04
2003:02
2007:04
Real Net Taxation
0.4
-0.8
1989:04
1998:04
2003:02
2007:04
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1989:04
1994:02
Real Government Consumption and
Investment less Real Net Taxation
1998:04
2003:02
2007:04
Debt
0.1
1
-0.1
0.5
-0.3
0
-0.5
-0.5
-0.7
-0.9
1989:04
1994:02
1998:04
2003:02
-1
2007:04
1989:04
25
1994:02
1998:04
2003:02
2007:04
Figure 3b
Moving Contemporaneous Cross Correlations with Real GDP: Labour Market Variables
Employment
Unemployment
1
0.4
0.8
0.2
0
0.6
-0.2
0.4
-0.4
0.2
-0.6
0
-0.2
1989:04
-0.8
1994:02
1998:04
2003:02
-1
1989:04
2007:04
1994:02
1998:04
2003:02
2007:04
Labour Productivity
1
0.8
0.6
0.4
0.2
0
1989:04
1994:02
1998:04
2003:02
2007:04
Real Average Weekly Earnings
Real Average Hourly Earnings
0.6
0.6
0.4
0.4
0.2
0.2
0
0
-0.2
-0.4
-0.2
-0.6
-0.4
-0.8
-0.6
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
-1
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
Real Labour Cost
Real Unit Labour Cost
0.8
0.2
0.6
0
0.4
-0.2
0.2
0
-0.4
-0.2
-0.6
-0.4
-0.8
-0.6
-0.8
1995:02
1997:04
2000:02
2002:04
2005:02
-1
1995:02 1997:04 2000:02 2002:04 2005:02 2007:04
2007:04
26
Figure 4
Moving Cross Correlations with Real GDP: Real Variables
Computed for the Phase Shift Exhibiting the Maximum Correlation
Government Expenditure : t+5
Government Consumption: t+5
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0
-0.2
-0.4
1989:04
1994:02
1998:04
2003:02
2007:04
-0.2
1989:04
1994:02
Government Investment: t+5
0.6
0.4
0.2
0
-0.2
1994:02
1998:04
2003:02
2007:04
0.1
-1E-15
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
1989:04
1994:02
1998:04
Total Imports: t+2
0.95
0.75
0.55
0.35
0.15
-0.05
1989:04
2003:02
2007:04
Net Exports Share: t+2
0.8
-0.4
1989:04
1998:04
1994:02
1998:04
27
2003:02
2007:04
2003:02
2007:04
Figure 4a
Moving Contemporaneous Cross Correlations with Real GDP: New Zealand Treasury Fiscal Variables
Computed for the Phase Shift Exhibiting the Maximum Correlation
Government Investment: t+5
Government Consumption: t+5
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0
-0.2
-0.2
-0.4
1989:04
0.8
1994:02
1998:04
2003:02
2007:04
-0.4
1989:04
0.4
0.6
0.2
0.4
0
0.2
-0.2
0
1998:04
2003:02
2007:04
-0.2
1989:04
0.4
0.2
0.2
0
0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.6
-0.8
1994:02
1998:04
2003:02
1994:02
-1
1989:04
2007:04
1994:02
Debt: t+3
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
-1
1989:04
2007:04
1998:04
2003:02
2007:04
Real Government Consumption and
Investment Less Real Net Taxation: t-1
Real Transfers: t-5
-0.8
1989:04
2003:02
1
0.8
1994:02
1998:04
Real Taxation: t+1
Real Government Consumption and
Investment: t+5
0.6
-0.4
1989:04
1994:02
1994:02
1998:04
28
2003:02
2007:04
1998:04
2003:02
2007:04
Figure 4b
Moving Contemporaneous Cross Correlations with Real GDP: Labour Market Variables
Computed for the Phase Shift Exhibiting the Maximum Correlation
Employment: t+2
Unemployment: t+1
1
0.2
0.8
0
0.6
-0.2
0.4
-0.4
0.2
-0.6
0
-0.8
-0.2
1989:04
1994:02
1998:04
2003:02
-1
1989:04
2007:04
1997:04
2000:02
2002:04
2005:02
2003:02
2007:04
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
Real Labour Cost: t+4
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
1995:02
1998:04
Real Average Hourly Earnings: t+2
Real Average Weekly Earnings: t+2
0.1
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
1991:03 1994:02 1997:01 1999:04 2002:03 2005:02 2008:01
1994:02
2007:04
29
Figure 5
Moving Percentage Standard Deviations from Trend: Prices and Monetary Variables
Terms of Trade
Nominal Trade Weighted Index
0.06
0.09
0.05
0.08
0.07
0.04
0.06
0.03
0.05
0.02
0.04
0.01
0
1989:04
0.03
1994:02
1998:04
2003:02
2007:04
0.02
1989:04
1994:02
2007:04
0.02
0.025
0.02
0.015
0.015
0.01
0.01
0.005
0.005
1994:02
1998:04
2003:02
0
1989:04
2007:04
Nominal 90-day Interest Rate
0.018
0.016
0.014
0.012
0.01
0.008
0.006
0.004
0.002
1989:04
2003:02
GDP Deflator
Consumer Price Index
0
1989:04
1998:04
1994:02
1998:04
2003:02
2007:04
Real 90-day Interest Rate
0.016
0.014
0.012
0.01
0.008
0.006
0.004
1994:02
1998:04
2003:02
2007:04
0.002
1989:04
1994:02
1998:04
2003:02
2007:04
M3(r)
5-Year Government Bond
0.035
0.014
0.03
0.012
0.01
0.025
0.008
0.02
0.006
0.015
0.004
0.002
1989:04
1994:02
1998:04
2003:02
2007:04
0.01
1989:04
30
1994:02
1998:04
2003:02
2007:04
DC(r)
0.03
0.025
0.02
0.015
0.01
0.005
1989:04
1994:02
1998:04
2003:02
2007:04
Figure 6
Moving Contemporaneous Cross Correlations with Real GDP: Prices and Monetary Variables
Nominal Trade Weighted Index
Terms of Trade
0.8
0.75
0.7
0.6
0.25
0.5
0.4
-0.25
0.3
-0.75
1989:04
1994:02
1998:04
2003:02
2007:04
0.2
1989:04
1994:02
0.6
0.4
0.2
-1E-15
-0.2
-0.4
1994:02
1998:04
2003:02
2007:04
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
1989:04
2007:04
1994:02
1998:04
2003:02
2007:04
Real 90-day Interest Rate
Nominal 90-day Interest Rate
0.7
0.5
0.3
0.1
-0.1
-0.3
-0.5
1989:04
2003:02
GDP Deflator
Consumer Price Index
-0.6
1989:04
1998:04
0.7
0.5
0.3
0.1
-0.1
-0.3
1994:02
1998:04
2003:02
2007:04
-0.5
1989:04
31
1994:02
1998:04
2003:02
2007:04
M3(r)
5-Year Government Bond
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.5
1989:04
1994:02
1998:04
2003:02
2007:04
-0.4
1989:04
1994:02
1998:04
DC(r)
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
-0.8
1989:04
1994:02
1998:04
32
2003:02
2007:04
2003:02
2007:04
Figure 7
Moving Cross Correlations with Real GDP : Prices and Monetary Variables
Computed for the Phase Shift Exhibiting the Maximum Correlation
Consumer Price Index: t+5
1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
1989:04
GDP Deflator: t+5
1
0.6
0.8
0.4
0.6
0.2
0.4
0
0.2
1994:02
1998:04
2003:02
0
1989:04
2007:04
1
0.8
0.6
0.4
0.2
1994:02
1998:04
2003:02
1994:02
2007:04
1
0.8
0.6
0.4
0.2
0
-0.2
-0.4
1989:04
1994:02
2007:04
1998:04
2003:02
2007:04
1998:04
2003:02
2007:04
DC(r): t+5
M3(r): t+5
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0
-0.2
-0.2
-0.4
1989:04
2003:02
5-Year Government Bond: t+3
Real 90-day Interest Rate: t+3
0
1989:04
1998:04
Nominal 90-day Interest Rate: t+3
0.8
-0.2
1989:04
1994:02
1994:02
1998:04
2003:02
2007:04
-0.4
1989:04
33
1994:02
1998:04
2003:02
2007:04
C
Variables and their Sources
All series were obtained from SNZ‟s Infoshare website, except for those containing the source reference
„RBNZ‟, and except for fiscal variables sourced from the New Zealand Treasury‟s fiscal data set for the
time period 1982q2 to 2010q4. Where appropriate, notes are included on the calculation of variables that
were not directly available in raw data form.
SA = seasonally adjusted, Log = Log transformed
Output and expenditure variables
1.
Gross Domestic Product – Expenditure Measure
Prodn A/C, GDP & GDE, Constant Prices, SA, Qrtly ($Millions) [Log]
2.
Private Non Profit Organisations and Households Combined
Use of Inc, Final Consumption Exp. Govt/Pvte Sectors, Constant Prices, SA, Qrtly ($Millions) [Log]
3.
Non Durable Goods
Use of Inc, Final Consumption Exp. by HH Item Type, Constant Prices, SA, Qrtly ($Millions) [Log]
4.
Durable Goods
Use of Inc, Final Consumption Exp. by HH Item Type, Constant Prices, SA, Qrtly ($Millions) [Log]
5.
Private Gross Fixed Capital Formation
Capital A/C, Gross Fixed Cap. Form. by Sect/O'ship, Constant Prices, SA, Qtrly ($Millions) [Log]
6.
Residential Buildings
Capital A/C, Gross Fixed Capital Formation by Asset, Constant Prices, SA, Qtrly ($Millions) [Log]
7.
Non Residential Buildings
Capital A/C, Gross Fixed Capital Formation by Asset, Constant Prices, SA, Qtrly ($Millions) [Log]
8.
Other
[Private Gross Fixed Capital Formation – (Residential Buildings + Non Residential Buildings)]
($Millions) [Log]
9.
Government Expenditure
[Government Consumption + Government Investment] ($Millions) [Log]
10. Government Consumption
Use of Inc, Final Consumption Exp. by Govt/Pvte Sectors, Constant Prices, SA, Qrtly ($Millions)
[Log] Note: Excludes the purchase of Frigates in 1997(2) and 1999(4)
11. Government Gross Fixed Capital Formation (Government Investment)
Capital A/C, Gross Fixed Cap. Form. by Sect/O'ship, Constant Prices, SA, Qrtly ($Millions) [Log]
12. Net Exports
[Total Exports – Total Imports] ($Millions) [Expressed as a share of GDP]
13. Total Exports
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
14. Exports of Goods
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
15. Exports of Services
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
34
16. Total Imports
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
17. Imports of Goods
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
18. Imports of Services
Components of External Transactions Account, Constant Prices, SA, Qrtly ($Millions) [Log]
Price and monetary variables
19. Terms of Trade
Overseas Trade Indexes, All Countries, Qrtly [Log]
20. Nominal Trade Weighted Index
Nominal and Real TWI, Figure 8b, RBNZ [Log]
21. Real Trade Weighted Index
Nominal and Real TWI, Figure 8b, RBNZ [Log]
22. Consumer Price Index (CPIXGST)
CPI All Groups for New Zealand, Qrtly, (Re-indexed to reflect GST-exclusive inflation rates) [Log]
23. Consumer Price Index Excluding Food, Fuel, and Government Purchases
Incomes and Prices, A3 Historical Series, RBNZ, Qrtly [Log]
24. GDP (expenditure) Deflator
Incomes and Prices, A3 Historical Series, RBNZ, Qrtly [Log]
25. Nominal 90-day Interest Rate
Interest rates, B2 Historical Series, RBNZ [Quarterly average]
26. Real 90-day Interest Rate
[((1+Nominal 90-day Interest Rate)/(1+Two-Year Ahead Inflation Expectation))*100]
Two-Year Ahead Inflation Expectation sourced from RBNZ
27. 5-Year Government Bond
Interest Rates, B2 Historical Series, RBNZ [Quarterly average]
28. 90-day Yield Gap
[Nominal 90-day Interest Rate – 5-Year Government Bond]
29. Floating First Mortgage New Customer Housing Rate
Interest Rates, B3 Historical Series, RBNZ [Quarterly average]
30. M3(r)
Money and Credit Aggregates, C1 Historical Series, RBNZ [Log]
31. DC(r)
Money and Credit Aggregates, C2 Historical Series, RBNZ [Log]
Other fiscal variables
All sourced from the New Zealand Treasury‟s fiscal data set, as previously used in Claus et al.(2006),
Dungey and Fry (2009), and Hall and McDermott (2011).
32. – 37.
Real government expenditure, real government consumption expenditure, real taxation, real
transfers, nominal gross government debt
35
Labour market variables
38. Employment
SNZ HLFS Persons employed in labour force („000), SA [Log]
39. Unemployment rate
SNZ HLFS Unemployment rate (%), Total, SA
40. Labour productivity
Production based real GDP/Employment [Log]
41. Real average weekly earnings
SNZ All industries total (ordinary time + overtime) average weekly earnings (FTE), deflated by
CPIXGST [Log]
42. Real average hourly earnings
RBNZ average hourly earnings (total), deflated by CPIXGST [Log]
43. Real labour cost
RBNZ Labour cost index (June 2009 = 1000), deflated by GDP deflator [Log]
44. Real unit labour cost
Real labour cost/production based real GDP [Log]
36
References
Backus, D. K., & Kehoe, P. J. (1992). International Evidence on the Historical Properties of Business
Cycles. The American Economic Review, 82(4), 864-888.
Backus, D. K., Kehoe, P. J., & Kydland, F. E. (1992). International Real Business Cycles. Journal of Political
Economy, 100(4), 745-775.
Backus, D. K., Kehoe, P. J., & Kydland, F. E. (1994). Dynamics of the Trade Balance and the Terms of
Trade: The J-Curve? The American Economic Review, 84(1), 84-103.
Baxter, M., & King, R. G. (1999). Measuring Business Cycles: Approximate Band-Pass Filters for
Economic Time Series. Review of Economics and Statistics, 81(4), 575-593. doi:
10.1162/003465399558454
Burns, A. F., & Mitchell, W. C. (1946). Measuring Business Cycles: National Bureau of Economic Research,
Inc.
Chetwin, W. (2012). Business cycle review, 1998-2011. Reserve Bank of New Zealand Bulletin, 75(1), March,
14-27.
Choy, K. M. (2011). Business cycles in Singapore: stylised facts for a small open economy. Pacific Economic
Review, 16(1), 18-35. doi: 10.1111/j.1468-0106.2010.00532.x
Christiano, L. J., & Eichenbaum, M. (1992). Current Real-Business-Cycle Theories and Aggregate LaborMarket Fluctuations. The American Economic Review, 82(3), 430-450.
Claus, I., Gill, A., Lee, B., &McLellan, N. (2006). An empirical investigation of fiscal policy in New
Zealand. New Zealand Treasury Working Paper 06/08.
Crosby, M. and Otto, G. (1995). Real business cycle models: An introduction. The Australian Economic
Review, Third Quarter, 55-70.
Diebold, F. X., & Rudebusch, G. D. (1996). Measuring Business Cycles: A Modern Perspective. The
Review of Economics and Statistics, 78(1), 67-77.
Dungey, M. &Fry, R. (2009). The Identification of Fiscal and Monetary Policy in a Structural VAR.
Economic Modelling, 26(6), November, 1147-1160.
Fisher, L. Otto, G. and Voss, G. (1996). Australian business cycle facts. Australian Economic Papers, 35 (67),
300-320.
37
Friedman, M. (1957). A Theory of the Consumption Function: National Bureau of Economic Research, Inc.
Greenwald, B. C., Stiglitz, J. E., Hall, R. E., & Fischer, S. (1988). Examining Alternative Macroeconomic
Theories. Brookings Papers on Economic Activity, 1988(1), 207-270.
Hall, V.B., Kim, K, & Buckle, R. A. (1998). Pacific Rim business cycle analysis: synchronisation and
volatility. New Zealand Economic Papers, 32(2), 129-159.
Hall, V. B. & McDermott, C. J. (2009). The New Zealand business cycle. Econometric Theory, 25,1050-1069.
Hall, V. B. & McDermott, C. J. (2011). Unobserved components business cycles for New Zealand. What
are they, and what might drive them? The Economic Record, 87(277), June, 294-317.
Harvey, A. (1985). Trends and Cycles in Macroeconomic Time Series. Journal of Business and Economic
Statistics, 3(3), 216-227.
Harvey, A. (1989). Forecasting, Structural Time Series Models and the Kalman Filter. Cambridge: Cambridge
University Press.
Harvey, A. C., & Jaeger, A. (1993). Detrending, Stylized Facts and the Business Cycle. Journal of Applied
Econometrics, 8(3), 231-247.
Haywood, E. (1972). The Dating of Post War Business Cycles in New Zealand: 1946-70. Reserve Bank of New
Zealand Research Paper 4, 15.
Haywood, E. & Campbell, C. (1976). The New Zealand Economy: Measurement of Economic Fluctuations and
Indicators of Economic Activity, 1947-74. Reserve Bank of New Zealand Research Paper 19, 31.
Hodrick, R. J., & Prescott, E. C. (1980). Postwar U.S. Business Cycles: An Empirical Investigation
Discussion Papers: Carnegie-Mellon University.
Kim, K., Buckle, R. A., & Hall, V. B. (1994). Key Features of New Zealand Business Cycles. Economic
Record, 70(208), 56-72. doi: 10.1111/j.1475-4932.1994.tb01825.x
King, R. G., & Rebelo, S. T. (2000). Resuscitating Real Business Cycles NBER Working Paper Series (Vol.
w7534).
Krugman, P. R., Baldwin, R. E., Bosworth, B., & Hooper, P. (1987). The Persistence of the U.S. Trade
Deficit. Brookings Papers on Economic Activity, 1987(1), 1-55.
Kydland, F., & Prescott, E. (1990). Business Cycles: Real Facts and a Monetary Myth. Federal Reserve Bank
of Minneapolis Quarterly Review, 14(2), 3-18. doi: citeulike-article-id:909451
38
Lucas, R. E. (1977). Understanding business cycles. Carnegie-Rochester Conference Series on Public Policy, 5(1), 729.
McCaw, S. (2007). Stylised facts about New Zealand business cycles. Reserve Bank of New Zealand Discussion
Paper, DP2007/04.
McKelvie, S. (2011). Stylised facts of New Zealand business cycles: a post-1987 perspective. BCA
Honours Research Project, School of Economics & Finance, Victoria University of Wellington,
October, 30.
Ramey, G., & Ramey, V. A. (1995). Cross-country evidence on the link between volatility and growth.
American Economic Review, 85, 1138-1151.
Reddell, M. & C. Sleeman (2008). Some perspectives on past recessions. Reserve Bank of New Zealand
Bulletin, 71(2), 5-21.
Sanyal, A., & Ward, B. D. (1995). Has a structural break affected New Zealand business cycles? Department
of Economics and Marketing discussion paper: Lincoln University.
Tawadros, G. B. (2011). The stylised facts of Australia's business cycle. Economic Modelling, 28(1-2), 549556.
39